I'm working on a significant refactor in the branch flat-files. The goals of this refactor are as follows:
- be smarter about the python object that comes out of pyfec. Right now it's a little haphazard where fields are located, and it has that terrible flat_filing key. I'll try to give things better names and put them in more logical places. These changes will be breaking whenever merged back in
- Create a way to dump the transactions from a filing into csvs for easy import to a database via file copying. This part should be irrelevant to current users.
I don't believe that pyfec has a mailing list. If you're a user who wants to be informed about potentially breaking changes, please contact rachel.shorey@nytimes.com. When I merge in the new branch, I'm going to upgrade to v1 and will be good about version tagging from there on out.
Really. On a given day, we make no promises that it works. We make no promises that it's not wrong. We make no promises that we won't push breaking hotfixes. If you're going to use it, we strongly recommend a fork.
pip install -e git+git@github.com:newsdev/nyt-pyfec.git#egg=nyt-pyfec
python -m pyfec.demo
...if you, like me, are other than awesome at git.
- Make the updates in the fec-csv-sources repo, commit and push to github
- Go your local nyt-pyfec repo, cd into pyfec
- Make sure everything is backed up and committeed and stuff
- If you've done this before,
git submodule update --remote fec-csv-sources - If you haven't done it before,
rm -r fec-csv-sources && git submodule update --init --remote fec-csv-sources - Commit and push
- Voila?